UK Government Spending

graph layouts

Photo by Taylor Wilcox on Unsplash

Ingest

supplier to recipient data

nodes <- read.csv("archetypes/uk-government-spending/nodes.csv", header = TRUE, stringsAsFactors = FALSE)
nodes$color <- as.character(paste0("#", nodes$color))
head(nodes, n = 10)

edges <- read.csv("archetypes/uk-government-spending/edges.csv", header = TRUE, stringsAsFactors = FALSE)
edges$color <- as.character(paste0("#", edges$color))
head(edges, n = 10)

Wrangle

transform to graph structure

# Convert data to graph data frame
# directed=true means it is a flow not just an association between nodes
g <- graph.data.frame(edges, directed=TRUE, vertices=nodes)
g

## IGRAPH 07e20f0 DN-- 107 106 -- 
## + attr: name (v/c), billion_pounds (v/n), percent (v/n),
## | contribution_per_taxpayer_per_day (v/n), color (v/c), billion_pounds
## | (e/n), percent (e/n), contribution_per_taxpayer_per_day (e/n), color
## | (e/c)
## + edges from 07e20f0 (vertex names):
## [1] Budget                  ->Incomes                 
## [2] Incomes                 ->Taxes on Wealth & Income
## [3] Taxes on Wealth & Income->Income Tax              
## [4] Income Tax              ->Paye                    
## [5] Income Tax              ->Self Employed           
## + ... omitted several edges

Ensuring proper attributes for Edges

Computing Edges Weights (as per number of students flow)

df_graph <- g %>% 
   as_tbl_graph() %>% 
    activate(edges) %>% 
    # computes numbers between 0 and 1 according to Number of Student Number
    mutate(edge_weights = log10(billion_pounds)/10)
df_graph

## # A tbl_graph: 107 nodes and 106 edges
## #
## # A rooted tree
## #
## # Edge Data: 106 x 7 (active)
##    from    to billion_pounds percent contribution_per_taxpay~ color edge_weights
##   <int> <int>          <dbl>   <dbl>                    <dbl> <chr>        <dbl>
## 1     1     2          753.  1                              0 #4e1~       0.288 
## 2     2     3          221   0.294                          0 #944~       0.234 
## 3     3     4          177.  0.236                          0 #944~       0.225 
## 4     4     5          146   0.194                          0 #944~       0.216 
## 5     4     6           24.3 0.0323                         0 #944~       0.139 
## 6     4     7            7   0.00930                        0 #944~       0.0845
## # ... with 100 more rows
## #
## # Node Data: 107 x 5
##   name                     billion_pounds percent contribution_per_taxpay~ color
##   <chr>                             <dbl>   <dbl>                    <dbl> <chr>
## 1 Budget                             753.   1                            0 #4e1~
## 2 Incomes                            753.   1                            0 #4e1~
## 3 Taxes on Wealth & Income           221    0.294                        0 #944~
## # ... with 104 more rows

Layout & Aesthetic Parameters

Choose the types of Edges you would like to display

## Edge Types
# geom_edge_arc
# geom_edge_arc0
# geom_edge_arc2
# geom_edge_density
# geom_edge_diagonal
# geom_edge_diagonal0
# geom_edge_diagonal2
# geom_edge_elbow
# geom_edge_elbow0
# geom_edge_elbow2
# geom_edge_fan
# geom_edge_fan0
# geom_edge_fan2
# geom_edge_hive
# geom_edge_hive0
# geom_edge_hive2
# geom_edge_link
# geom_edge_link0
# geom_edge_link2

## Layouts can be computationally intensive
# Uncomment only for use

##################
# igraph layouts #
##################

# layout <- create_layout(df_graph, layout = 'igraph', algorithm = 'bipartite')
# layout <- create_layout(df_graph, layout = 'igraph', algorithm = 'circle')
# layout <- create_layout(df_graph, layout = 'igraph', algorithm = 'dh')
# layout <- create_layout(df_graph, layout = 'igraph', algorithm = 'drl')
# layout <- create_layout(df_graph, layout = 'igraph', algorithm = 'fr')
# layout <- create_layout(df_graph, layout = 'igraph', algorithm = 'gem')
# layout <- create_layout(df_graph, layout = 'igraph', algorithm = 'graphopt')
# layout <- create_layout(df_graph, layout = 'igraph', algorithm = 'grid')
layout <- create_layout(df_graph, layout = 'igraph', algorithm = 'kk')
# layout <- create_layout(df_graph, layout = 'igraph', algorithm = 'lgl')
# layout <- create_layout(df_graph, layout = 'igraph', algorithm = 'mds')
# layout <- create_layout(df_graph, layout = 'igraph', algorithm = 'nicely')
# layout <- create_layout(df_graph, layout = 'igraph', algorithm = 'sphere')
# layout <- create_layout(df_graph, layout = 'igraph', algorithm = 'star')
# layout <- create_layout(df_graph, layout = 'igraph', algorithm = 'sugiyama')

## graphlayouts
# layout <- create_layout(df_graph, layout = "nicely")
# layout <- create_layout(df_graph, layout = "stress")
# layout <- create_layout(df_graph, layout = "focus", focus = 1)

View

graph

theme_opts <- theme(
  legend.position = "none",
  legend.title = element_blank(),
  axis.text = element_blank(),
  axis.line = element_blank(),
  axis.ticks = element_blank(),
  axis.title = element_blank(),
  panel.grid = element_blank(),
  panel.grid.major = element_blank(),
  panel.grid.minor = element_blank(),
  panel.background=element_rect(fill="white", colour="white"),
  panel.border = element_blank(),
  plot.background = element_blank(),
)

colors <- unique(E(df_graph)$color)
color_palette <- c( colors )
color_palette

##  [1] "#4e1c12" "#944b36" "#5f0801" "#b37b65" "#999999" "#e6e6e6" "#9acdaa"
##  [8] "#46a49d" "#6ab6ad" "#56827a" "#87a38e" "#326965" "#7f7f7f"

v1 <- ggraph(layout) + 
  geom_edge_elbow(aes(edge_width = edge_weights, color = color)) + # aes(edge_width = edge_weights)
  geom_node_point(aes(size=billion_pounds, color = color)) +
  geom_node_label( aes(label=name, fill = color), color = "white", repel = FALSE, vjust = -0.5, hjust = 0.5, size = 4, family = "inconsolata") +
  scale_size_continuous(range = c(1, 12)) + 
  scale_color_manual(values = color_palette) +
  coord_fixed(clip = 'off') + 
  theme_graph()+
  theme_opts

girafe(ggobj = v1, width_svg = 16, height_svg = 14,
       options = list(opts_sizing(rescale = TRUE, width = 1.0)))

References

citations for narrative and data sources

Source: